Abstract: To obtain reliable dynamical models of intracellular gene regulation directly from expression data, one, to get the structure of the model, needs to first infer existing causal influences among the genes of interest, i.e. solve the network inference problem. Network inference is thus an enabling technology and key problem in Systems biology. Through systematic perturbation experiments, utilizing e.g. siRNA knock-downs, and measurement of the resulting gene expression changes, using e.g. qRT-PCR, followed by network inference, one could in theory test thousands of genetic interaction hypotheses in one set of experiments.

Recent scrutiny of promising studies, such as the SOS pathway in E. coli by Gardner et al. in Science (2003) and Snf1 pathway in S. cerevisiae by Lorenz et al. in PNAS (2009), has shown that the inferred models are unreliable. Out of 35 inference methods benchmarked in the DREAM challenge only some yielded networks more similar to the true network than expected when picking links randomly.

The main aims of this project is to (i) develop a method for finding the most likely network based on partly informative data, (ii) find the best inference method for a variety of network and data properties, (iii) develop a parallel algorithm for robust network inference, (iv) explore the limits for what can be inferred from different types of experiments, (v) develop a method for iterative experiment design, and (vi) apply these methods to qRT-PCR data from in vitro experiments in epidermoid squamous cell carcinoma cell line A431.

Relative to the granted budget and anticipated main results, i.e. SCI journal articles, we have exceeded the plan by 750%. We were granted 15% of the requested budget and despite this managed to produce 9 journal articles (currently 4 published, 4 submitted, and 1 is almost ready for submission). Due to the budget cut, we have not been able to address aim (iii) and (v) above. It remains for future work. I would like to highlight the value of the international collaboration with Prof. Erik Sonnhammer, which has enabled me to work with him and two of his PhD students, thus in part compensating for the shortage of PhD students in Taiwan. Also in collaboration with Prof. Mika Gustafsson at Linköping University we participated in the DREAM10 Respiratory Viral DREAM Challenge. I have also taught one course N181100 MODEL SELECTION AND NETWORK INFERENCE OF COMPLEX SYSTEMS in the fall 2016 and 2017 with a total of 15 master and PhD students attending. This course cover the main problem addressed in this research project, i.e. model selection and network inference. A total of 24 students have thus learned about network inference as a result of this project.

Members: Rain wu, Lewis Hsu, Paul Tsai, Tim Hsia, Torbjörn Nordling